Unseen burden: Prevalence and determinants of possible sarcopenia in Indonesian older adults – a secondary data analysis

Abstract

Background

Indonesia is experiencing rapid population aging, raising concern about muscle weakness in later life; possible sarcopenia offers early identification through low handgrip strength.

Objective

To estimate the prevalence of possible sarcopenia and its determinants among older adults in Indonesia.

Methods

This secondary analysis used baseline data from the Indonesian Longitudinal Aging Survey 2023 and included adults aged sixty years or older with valid handgrip measurements. Possible sarcopenia followed the Asian Working Group for Sarcopenia 2019 thresholds using the maximum of two trials per hand. Covariates included sociodemographic, behavioral, functional, and clinical factors. We applied descriptive statistics, bivariate tests, and multivariable logistic regression, with probit average marginal effects. Sampling weights were unavailable in the public-use microdata; we treated enumeration areas as the primary sampling units, with available stratification and no weights; estimates were unweighted, and standard errors were design-based (survey-corrected).

Results

Among 1,598 participants, the prevalence of possible sarcopenia was 51.1%. Older age, low physical performance, and urban residence were associated with higher odds. Higher body mass index and better cognition were associated with lower odds. A prespecified sex-by-body mass index interaction suggested attenuation of the protective association of body mass index among women.

Conclusion

Possible sarcopenia is common among older Indonesians and is patterned by age, body composition, cognition, functional status, and residential context. Community health services, including community nursing services, can integrate routine handgrip assessment with brief interventions on strength activities, nutrition, and cognitive engagement to identify risk early and inform preventive care.

Background

Population aging is a global phenomenon that is reshaping social and economic structures, including in Asia and the Pacific (Kazeminia et al., 2020). This trend, marked by a growing proportion of older individuals, raises important concerns for public health and quality of life. Health systems, therefore, need to reassess geriatric care models and develop targeted interventions for age-related conditions (Pitchalard et al., 2022).

Within low and middle-income countries, the health consequences of population aging are multifaceted. Older adults are experiencing a rise in chronic noncommunicable conditions, partly because adverse early life conditions such as malnutrition and childhood illness have effects that persist into adulthood (Smith et al., 2021; Zolnikov, 2015). Indonesia exemplifies this transition, with a rapidly expanding older population and growing public health needs, including age-related morbidities such as sarcopenia (Irwan et al., 2024). This trajectory, together with Indonesia’s position as the country with the largest older population in Southeast Asia, underscores the urgency of addressing age-related health issues (Kandinata et al., 2023).

Sarcopenia is a geriatric syndrome characterized by progressive and generalized loss of skeletal muscle mass and strength that leads to physical disability, poorer quality of life, and increased risk of mortality (Altaf et al., 2024). The Asian Working Group for Sarcopenia introduced possible sarcopenia in 2019 as an accessible first step that uses simple measures such as calf circumference to enable early screening in resource-limited settings and reduce progression (Kim et al., 2025; Won, 2023). Compared with confirmed sarcopenia, which requires more complex examinations, this approach is intended to facilitate early detection and timely intervention in primary care and community settings, including nursing community services (Crosignani et al., 2021; Shin et al., 2022).

The study of possible sarcopenia in Indonesia remains limited. A study conducted in Surabaya reported poor diagnostic performance of these modalities, indicating the need for multicenter research across different regions to confirm the findings (Kandinata et al., 2023). A nationwide study is also needed to establish optimal cutoff values for calf circumference, muscle strength, and physical performance that are specific to the Indonesian population (Setiati et al., 2025).

Accordingly, this study investigated the prevalence and determinants of possible sarcopenia among older adults in Indonesia using data from the first wave of the Indonesian Longitudinal Aging Survey (ILAS) 2023. This study aims to translate sarcopenia from an abstract clinical construct into a measurable public health target. Recognizing that nurses in community health centers are uniquely positioned to serve as the frontline for geriatric screening, this research seeks to demonstrate how simple tools, such as handgrip strength measurements, can be incorporated into routine assessments. By doing so, nursing professionals can facilitate early identification and initiate timely interventions to prevent functional decline. Identification of key risk factors and vulnerable subgroups is intended to inform targeted screening and interventions, thereby maximizing impact and guiding resource allocation.

Methods

Study Design and Data Sources

This secondary analysis used data from the first wave of ILAS 2023. ILAS is a population-based longitudinal survey that collects demographic, socioeconomic, and health information, including objective physical measurements, among older Indonesians. Fieldwork used computer-assisted personal interviewing (CAPI) from May to June 2023. ILAS 2023 enrolled 4,084 respondents aged ≥45 years, selected via multistage random sampling from nine provinces.

The analytic sample was restricted to adults aged ≥60 years. Because sampling weights were not available in the public-use microdata, estimates were generated without weights; the multistage design was accommodated by treating enumeration areas as primary sampling units and using available stratification in the survey estimation framework, so that standard errors reflect clustering. All confidence intervals, odds ratios, and average marginal effects were produced from survey-corrected models.

Participants

Eligibility required age ≥60 years at interview and a valid handgrip strength measurement. For multivariable and interaction analyses, only observations with complete data on the outcome and all covariates were included (complete case analysis).

Definition of Possible Sarcopenia

The Asian Working Group for Sarcopenia 2019 (AWGS 2019) criteria were applied, defining possible sarcopenia as low muscle strength only: maximal handgrip <28 kg for men and <18 kg for women. Trained field staff obtained two trials for each hand under a standardized protocol; the highest value across the four trials was used. Device make/model information was not available in the ILAS report or codebook.

Socio-Demographic Variables

Variables included age (years), sex, education (<primary, primary, junior high, senior high, higher), marital status (married/cohabiting; single; divorced/widowed), residence (urban/rural), health insurance ownership (Have/Not Have), and economic status (tertiles of a proxy for monthly household expenditure: low, moderate, high).

Health Behaviors and Physical Performance

Smoking and alcohol consumption were categorized as never, current, or former. Physical activity was coded as active versus inactive. Because instrumented performance tests were unavailable, low physical performance was proxied by self-reported great difficulty or inability to walk or climb stairs; this proxy is a pragmatic choice given secondary data constraints.

Clinical Aging-Related Conditions

Predictors included: (i) cognitive function assessed using the Six-Item Screener (SIS); (ii) depressive symptoms measured with the CES-D-10; (iii) comorbidity count (sum of diagnosed chronic conditions); and (iv) functional limitations (count of reported difficulties across the five Washington Group domains).

Statistical Analysis

Baseline characteristics and the prevalence of possible sarcopenia were summarized using descriptive statistics (means/SD or medians/IQR for continuous variables; counts/percentages for categorical variables). Between-group comparisons used Pearson’s χ² tests for categorical variables and t-tests or ANOVA for continuous variables, implemented within the survey estimation framework to obtain design-based standard errors. Determinants of possible sarcopenia were examined using survey-corrected binary logistic regression, with adjusted odds ratios (ORs) reported with 95% confidence intervals (CIs).

Average marginal effects (AMEs) were estimated from a probit model to express absolute percentage-point changes in predicted probability; probit was selected over logistic AME because it yields numerically stable derivatives across the probability range and produces probability-scale effects that are essentially equivalent yet more directly interpretable while avoiding the non-collapsibility limitations of odds ratios.

Five sex interaction models were prespecified to assess effect modification. Complete case analysis was applied. Prior to modeling, missingness patterns were profiled across age, sex, residence, and expenditure tertiles; no strong systematic gradients were observed, supporting a plausible Missing at Random assumption, though residual bias cannot be excluded. Multicollinearity in the final models was assessed using variance inflation factors (VIFs), with no concerning multicollinearity detected. Statistical significance was set at α=0.05 (two-sided). Analyses were conducted in Stata version 16.0.

Ethical Consideration

The application for ethical clearance and a research permit for ILAS 2023 was submitted to the National Research and Innovation Agency (Badan Riset dan Inovasi Nasional). The Ethics Commission for Social and Humanities Research subsequently granted ethical clearance under reference number 558/KE.01/SK/12/2022. This approval confirms that the research adheres to established ethical standards for social and humanities studies.

Results

Study Sample

The analysis included 1,771 adults aged ≥60 years. After excluding participants with missing handgrip strength data (n = 173, 9.8%), the final analytic sample comprised 1,598 individuals. Of these, 817 met the criteria for possible sarcopenia, yielding a prevalence of 51.1%.

Baseline Characteristics

Compared with those without possible sarcopenia, cases were older, had lower BMI, and demonstrated markedly weaker handgrip strength. Comorbidity counts increased, functional limitations and low physical performance became more common, with women making up a larger share, and marital status distributions differed. No differences were found in residence, health insurance, economic status, smoking, alcohol use, or physical activity (Table 1).

Table 1.

Baseline characteristics by possible sarcopenia status (n = 1,598)

Characteristic	No Sarcopenia (n = 781)	Sarcopenia (n = 817)	p
Age, years, M ± SD	65.7 ± 5.1	70.3 ± 7.9	< 0.001
Age group, n (%)			< 0.001
60–70	655 (83.8)	486 (59.5)
71–80	114 (14.6)	233 (28.5)
≥ 81	12 (1.6)	98 (12.0)
BMI (kg/m2), M ± SD	23.4 ± 4.5	22.3 ± 4.7	< 0.001
BMI group, n (%)
Underweight	101 (12.9)	160 (19.6)
Normal	431 (55.2)	453 (55.5)
Overweight	184 (23.6)	155 (19.0)
Obese	65 (8.3)	49 (6.0)
Handgrip strength (kg), M ± SD	27.8 ± 7.1	15.7 ± 5.8	< 0.001
Comorbidities, n (%)			0.020
0–2	544 (69.7)	516 (63.1)
3–4	154 (19.7)	202 (24.7)
≥ 5	83 (10.6)	99 (12.2)
Depression score (CES-D), M ± SD	3.3 ± 3.8	2.8 ± 4.8	0.023
Cognitive score (0–6), M ± SD	4.2 ± 1.6	3.3 ± 1.9	< 0.001
Sex, n (%)			0.032
Female	399 (51.1)	461 (56.4)
Male	382 (48.9)	356 (43.6)
Education, n (%)			0.102
No education	90 (11.5)	122 (15.0)
Primary	455 (57.9)	471 (58.0)
Junior high	90 (11.4)	85 (10.5)
Senior high	85 (10.8)	84 (10.3)
Higher education	66 (8.4)	50 (6.2)
Marital status, n (%)			< 0.001
Single	10 (1.3)	19 (2.3)
Married	534 (68.4)	439 (53.7)
Separated	237 (30.3)	369 (44.6)
Smoking status, n (%)			0.110
Never	462 (59.1)	509 (62.3)
Current	210 (26.9)	183 (22.4)
Former	109 (14.0)	125 (15.3)
Alcohol use, n (%)			0.062
Never	741 (94.9)	781 (95.6)
Current	8 (1.0)	2 (0.3)
Former	32 (4.1)	34 (4.2)
Residence, n (%)			0.107
Urban	331 (46.6)	450 (50.7)
Rural	379 (53.4)	438 (49.3)
Insurance ownership, n (%)			0.449
Yes	583 (74.6)	596 (72.9)
No	198 (25.4)	221 (27.1)
Economic status, n (%)			0.148
Low	247 (31.6)	295 (36.1)
Moderate	274 (35.1)	260 (31.8)
High	260 (33.3)	262 (32.1)
Functional limitation score, M ± SD	0.6 ± 0.9	1.0 ± 1.2	< 0.001
Functional limitation group, n (%)
No limitation	451 (57.8)	370 (45.3)
Some limitation	292 (37.4)	338 (41.4)
Severe limitation	38 (4.8)	109 (13.3)
Physical performance, n (%)			< 0.001
Normal	604 (77.3)	540 (66.1)
Some difficulty	112 (14.3)	139 (17.0)
Low performance	65 (8.3)	138 (16.9)

Note. Continuous variables are presented as means and standard deviations (M ± SD); categorical variables are presented as frequencies and percentages. BMI = body mass index; CES-D = Center for Epidemiologic Studies Depression Scale. p-values represent group comparisons between participants with and without sarcopenia, based on independent-samples t-tests for continuous variables and chi-square tests for categorical variables.

Multivariable Associations (Logistic Regression)

In fully adjusted models (n = 1,594; four participants were excluded due to missing covariate data), older age, low physical performance, and urban residence were associated with higher odds of possible sarcopenia, whereas higher BMI and higher cognitive scores were protective. Although women constituted a larger share of cases in bivariate comparisons, the association attenuated and was not significant after multivariable adjustment. Likewise, the bivariate difference in depressive symptoms was no longer significant after adjustment. Relative to no formal education, senior high school education showed higher odds. Non-significant covariates were sex, primary/junior-high/higher education (vs <primary), marital status, smoking status, alcohol use, health insurance, economic status, physical activity, comorbidity count, functional limitation count, and depressive symptoms (Table 2).

Table 2.

Multivariable logistic regression for possible sarcopenia

Predictor	OR (95% CI)	p
Female	1.10 (0.79–1.53)	0.577
Age (per year)	1.09 (1.07–1.11)	< 0.001
BMI (per kg/m2)	0.97 (0.94–0.99)	0.013
Education
Primary	1.38 (0.97–1.96)	0.074
Junior high	1.52 (0.94–2.45)	0.090
Senior high	2.06 (1.24–3.43)	0.005
Higher education	1.65 (0.94–2.93)	0.084
Marital status
Single	2.01 (0.87–4.63)	0.100
Divorced/Widowed	1.26 (0.98–1.63)	0.075
Comorbidity count (per additional condition)	1.04 (0.99–1.09)	0.087
Smoking status
Never	1.04 (0.73–1.48)	0.832
Former	1.01 (0.70–1.47)	0.955
Alcohol use
Never	1.77 (0.44–7.16)	0.422
Former	1.49 (0.33–6.67)	0.606
Low physical performance	1.45 (1.00–2.10)	0.048
Functional limitations (per unit increase)	1.08 (0.96–1.22)	0.203
Depression score (per unit increase)	1.00 (0.97–1.03)	0.872
Cognitive score (per unit increase)	0.81 (0.75–0.87)	< 0.001
Urban residence	1.43 (1.12–1.82)	0.004
Insurance (yes)	0.88 (0.68–1.14)	0.327
Economic status
Medium	0.94 (0.72–1.22)	0.636
High	0.90 (0.69–1.19)	0.464
Physical activity (active)	0.85 (0.68–1.06)	0.148

Note. OR = odds ratio; CI = confidence interval; BMI = body mass index. Odds ratios were obtained from multivariable logistic regression modeling possible sarcopenia as the dependent variable. Reference categories were male, no formal education, married/cohabiting, current smoker, current alcohol use, rural residence, uninsured, low economic status, and physically inactive. All variables were entered simultaneously in a fully adjusted model. Continuous predictors were modeled per 1-unit increase.

Marginal Effects (Probit Model)

Average marginal effects, estimated from a probit model, reinforced the logistic findings on an absolute probability scale: older age increased the probability of possible sarcopenia, whereas higher BMI and higher cognitive scores reduced it; low physical performance and urban residence were associated with a higher probability; and, relative to <primary education, senior-high education showed a higher probability. Non-significant AMEs were observed for sex, marital status, comorbidity count, functional limitation count, depression score, smoking status, alcohol use, health insurance, economic status (medium and high vs low), physical activity, and primary/junior-high/higher education vs <primary (Table 3).

Table 3.

Average marginal effects (AME) from probit model (n = 1,594)

Predictor	AME (dy/dx)	95% CI	p
Female	+0.019	−0.051 to +0.089	0.591
Age (per year)	+0.017	+0.014 to +0.022	< 0.001
BMI (per kg/m2)	−0.006	−0.012 to -0.001	0.015
Education
Primary	+0.066	−0.007 to +0.139	0.075
Junior high	+0.085	−0.015 to +0.185	0.094
Senior high	+0.152	+0.044 to +0.257	0.005
Higher	+0.106	−0.013 to +0.226	0.081
Marital status
Single	+0.146	−0.028 to +0.319	0.099
Divorced/Widowed	+0.050	−0.003 to +0.104	0.066
Comorbidity count (per additional condition)	+0.008	−0.001 to +0.018	0.104
Smoking status
Never	+0.007	−0.066 to +0.083	0.845
Former	−0.001	−0.079 to +0.077	0.978
Alcohol use
Never	+0.128	−0.161 to +0.420	0.389
Former	+0.091	−0.222 to +0.406	0.568
Low physical performance	+0.080	+0.003 to +0.157	0.042
Functional limitations (per unit increase)	+0.015	−0.009 to +0.040	0.229
Depression score (per unit increase)	+0.001	−0.004 to +0.006	0.778
Cognitive score (per unit increase)	−0.045	−0.059 to -0.030	< 0.001
Urban residence	+0.073	+0.022 to +0.124	0.005
Insurance (yes)	−0.025	−0.078 to +0.029	0.361
Economic status
Medium	−0.012	−0.069 to +0.042	0.648
High	−0.023	−0.082 to +0.033	0.425
Physical activity (active)	−0.033	−0.080 to +0.013	0.166

Note. AME = average marginal effect; CI = confidence interval; BMI = body mass index. AMEs were derived from a fully adjusted probit model including the same covariates as the logistic regression model in Table 2. Continuous predictors are expressed as a per-unit increase, as specified. Positive AME values indicate an increased probability of possible sarcopenia; negative values indicate a decreased probability.

Furthermore, Figure 1 illustrates the comparative magnitude of these determinants based on Average Marginal Effects. While advanced age is a highly significant predictor, its absolute impact (+1.79 percentage points per year) is relatively modest compared to structural and functional factors. Notably, socio-environmental determinants such as urban residence and low physical performance confer a substantial increase in probability (approximately 7-8 percentage points), comparable in magnitude to the protective benefit of a high cognitive score. Senior high education exhibits the largest estimated effect size (+15.26 percentage points), although the wider confidence interval indicates greater variability in this estimate compared to the high precision observed for age and cognition.

Figure 1.

Impact of key determinants on possible sarcopenia (AME expressed as percentage-point change in predicted probability)

Effect-Modification (Interaction Models)

Across five prespecified interaction models, adjusting for the full covariate set, only the sex-by-body mass index interaction was statistically significant; other sex interactions were non-significant. BMI’s protective association was evident in men but was substantially attenuated and statistically non-significant in women, with margins indicating a clear downward gradient in predicted probability for men and minimal change for women across the BMI range. Non-significant interactions were sex and physical activity, sex and comorbidity count, sex and cognitive score, and sex and depression score (Table 4).

Table 4.

Interaction models (sex as effect modifier; n = 1,594)

Interaction (Model)	Key Coefficients (OR, 95% CI, p)	Predicted Probabilities / Simple Effects
Sex × BMI	BMI (main effect): 0.91 (0.873-0.956), p < 0.001; Female × BMI: 1.08 (1.032-1.147), p = 0.002; Female (main effect): see model note (BMI centered)	Margins (BMI 15→30): Men 0.643→0.361; Women 0.521→0.502. Implied BMI OR in women ≈ 0.994 per kg/m2
Sex × Physical activity	Female × Active: 0.94 (0.608-1.457), p = 0.786; Active (main effect): 0.87 (0.632-1.215), p = 0.429	Margins: Men Inactive 0.512 vs Active 0.484; Women Inactive 0.538 vs Active 0.497
Sex × Comorbidity count	Female × Comorbidity: 0.94 (0.860-1.035), p = 0.220; Comorbidity (main effect): 1.08 (1.003-1.166), p = 0.042	Margins (0→5 conditions): Men 0.463→0.546; Women 0.512→0.534
Sex × Cognitive score	Female × Cognition: 1.09 (0.960-1.240), p = 0.181; Cognition (main effect): 0.77 (0.698-0.854), p < 0.001	Margins (0→6): Similar declines in both sexes (protective association of cognition)
Sex × Depression score	Female × Depression: 0.965 (0.914-1.018), p = 0.190; Depression (main effect): 1.022 (0.982-1.064), p = 0.285	Margins (CES-D range): Diverging but overlapping trajectories; interaction not statistically significant

Note. OR = odds ratio; CI = confidence interval. Each interaction model included sex, the focal predictor, their interaction term, and the full set of covariates shown in Table 2. BMI was mean centered prior to inclusion of the interaction term. Predicted probabilities were derived from survey-corrected marginal estimates.

Predicted probabilities from interaction models

Figure 2 displays sex-specific predicted probabilities from the interaction models. Panel A (Sex × BMI) shows a marked decline across BMI 15-30 among men (≈0.643 to 0.361) with a nearly flat pattern among women (≈0.521 to 0.502), consistent with a significant attenuation of BMI’s protective association in women. Panel B (Sex × Physical Activity) shows near-overlapping lines for inactive versus active status, indicating no meaningful sex-specific difference in the association with physical activity.

Figure 2.

Predicted probability of possible sarcopenia by sex across interacting variables

Panel C (Sex × Comorbidity) shows shallow, imprecise increases in predicted probability with higher comorbidity counts for both sexes, with wide confidence intervals and no evidence of differential slopes by sex. Panel D (Sex × Cognitive Score) shows parallel, negative slopes for both sexes across the SIS range, consistent with the main protective association of cognition and no sex modification. Panel E (Sex × Depression Score) shows largely flat, overlapping lines over the CES-D-10 range with wide intervals, again consistent with null interaction. Error bars denote 95% confidence intervals derived from the survey-corrected models

Discussion

Principal Findings

This study examined the prevalence and determinants of possible sarcopenia among older adults in Indonesia using data from the ILAS 2023. Possible sarcopenia is an early-identification construct intended for timely detection and intervention in primary care, employing simpler, more accessible criteria than those required for a definitive diagnosis of sarcopenia. The Asian Working Group for Sarcopenia (AWGS) 2019 introduced this concept to facilitate prompt recognition and management of individuals at risk, thereby supporting prevention-oriented care pathways (Kim et al., 2025; Shin et al., 2022; Ueshima et al., 2021).

Using handgrip strength as the sole proxy measure, the prevalence of possible sarcopenia in this cohort was 51.1%. This estimate is considerably higher than reports from other regional studies using comparable frameworks, which found prevalence rates of 38.5% to 46% in China (Chen et al., 2022; Wu et al., 2021), 32.2% in Thailand (Sanguankittiphan et al., 2023), and 2.9% in Japan (Miura et al., 2021) but remains lower than that reported in Taiwan, where the prevalence reached 68.7% (Wu et al., 2022). This disparity may be attributed to methodological differences; although these studies applied the AWGS 2019 standard, case ascertainment in other settings often included additional confirmatory procedures. For instance, assessments in China and Japan frequently combined muscle strength evaluation with the SARC-F questionnaire and the five-times chair-stand test, potentially yielding more precise classification and, consequently, a lower observed prevalence (Miura et al., 2021; Shin et al., 2022).

Consistent with established global evidence on age-related physiological decline, advancing age was identified as a primary determinant of possible sarcopenia (Chen et al., 2022; Kwon et al., 2025; Whaikid et al., 2025). This aligns with findings from longitudinal studies indicating that chronological aging independently drives muscle deterioration through mechanisms of anabolic resistance and cumulative cellular damage (Hendrickse et al., 2025; Pérez-Castillo et al., 2025).

Furthermore, low physical performance was associated with a higher probability of possible sarcopenia (OR 1.45, 95% CI 1.00-2.10, p = 0.048). Given that this value is near the threshold for statistical significance, the finding suggests a modest association that warrants further investigation with objective performance metrics rather than a definitive link. In contrast to the typical urban advantage hypothesis, urban residence was also associated with a significantly higher risk. These findings align with recent research suggesting that rapid urbanization in developing nations facilitates a transition to sedentary lifestyles and potentially unhealthy dietary patterns, thereby accelerating muscle loss (Boakye et al., 2023; Chen et al., 2024).

The analysis also highlighted a robust protective association between cognitive function and possible sarcopenia. This supports the muscle-brain axis hypothesis, which posits a bidirectional relationship where myokines released by healthy muscle tissue exert neuroprotective effects (Kostka et al., 2024; Rai & Demontis, 2022). Conversely, it is important to acknowledge the potential for reverse causality, in which cognitive impairment may precipitate sarcopenia by impairing motor planning and reducing engagement in physical activity (Ragusa et al., 2024; Zhang et al., 2023). This mutual influence suggests that cognitive and physical frailty should be viewed as interconnected rather than distinct geriatric syndromes.

Body mass index showed a protective association with possible sarcopenia, consistent with reports that underweight status and limited nutritional reserves undermine muscle strength in older Asian adults (Kubo et al., 2025; Mustafa & Singh, 2024). The body mass index by sex interaction indicated protection in men but not in women. In the predicted probability plots, the probability declined across the observed body mass index range among men, whereas it remained nearly flat among women, which aligns with evidence that the relation between obesity and sarcopenia differs by sex (Liu et al., 2023; Lu et al., 2024).

A distinctive finding of this study was the increased risk of possible sarcopenia among individuals with senior high school education compared to those with lower education. Rather than an anomaly, this inverse association likely reflects occupational history. Individuals in this cohort with lower formal education may have engaged in lifelong manual labor, thereby preserving muscle tone, whereas those with higher education were more likely to hold sedentary clerical roles (You et al., 2024). This suggests that in transitional economies, the physical demands in lower-status occupations, which may inadvertently offer protection against sarcopenia that educational attainment does not have.

In contrast to several meta-analyses reporting a strong correlation between depression and sarcopenia (Chang et al., 2017; Li et al., 2022; Zhu et al., 2025), this study found no significant association. This null result aligns with previous findings in South Korea (Byeon et al., 2016) and may be attributable to cultural reporting biases, where older Indonesians potentially stigmatize or underreport psychological distress. A conservative interpretation is warranted: CES-D items may emphasize somatic complaints that overlap with frailty, cultural stigma may depress reporting among older Indonesians, and cross-sectional data limit causal inference.

Implications for Practice and Policy

These findings support a stratified service model that links early detection to targeted management. At the community level, public health nurses should incorporate routine handgrip screening into Posyandu Lansia (integrated service post for older people) and other outreach posts to identify at-risk older adults. At primary care and geriatric services, nurses should connect positive screens to tailored packages that include progressive resistance exercise, protein and leucine-focused nutrition counseling, and referral to physical therapy. Attention is needed for women with higher body mass index, older adults with cognitive complaints, and educated yet sedentary groups, since health literacy alone is insufficient to prevent muscle loss.

Strengths and Limitations

The strengths of this study lie in its use of a large, nationally representative dataset from the ILAS 2023 and the application of rigorous statistical controls, including interaction terms, to explore subgroup heterogeneity. However, several limitations warrant consideration. The cross-sectional design prevents the establishment of causal pathways, particularly regarding the bidirectional nature of the cognition-sarcopenia relationship. Additionally, the definition of possible sarcopenia relied exclusively on handgrip strength, which aligns with AWGS screening algorithms but suggests that the lack of muscle mass data may lead to an overestimation of prevalence relative to clinical diagnostic standards. While handgrip strength was measured using a handheld dynamometer, the specific make and model were not documented in the ILAS 2023 public materials. Furthermore, this study employed a complete-case analysis under the assumption of “missing at random”. Although missingness was profiled across demographic groups, potential residual bias should be considered when interpreting the precision of these prevalence estimates. Finally, unmeasured confounders such as dietary protein intake and hormonal status may have influenced the observed associations.

Conclusion

Sociodemographic and functional determinants, including age, urban residence, education, cognitive status, and sex specific body composition, were associated with possible sarcopenia among older Indonesians. Integrating routine handgrip screening in community services with targeted nutritional, physical, and cognitive support may help mitigate functional decline across diverse risk groups. Nursing practice should prioritize holistic community-based care that links screening to individualized lifestyle counseling and clear referral pathways, including nutrition support and physical therapy. Future research should employ longitudinal designs to clarify causal pathways and to inform a national framework for sarcopenia prevention.

Funding Statement

Funding This study was supported by the Indonesia Endowment Fund for Education (Lembaga Pengelola Dana Pendidikan/LPDP). Recipient: Via Dolorosa Halilintar.

Declaration of Conflicting Interest

No conflict of interest to declare.

Author Contribution

VDH: study concept and development, data acquisition, data management and analysis, wrote the first and final draft of the manuscript. PJ: study concept and development, supervision, revised the final draft, provided technical input, and gave final approval of the version to be published.

Author Biography

Via Dolorosa Halilintar, M.Pharm, MPH, is an Academia and Consultant for Development Studies. His Research interests are in Socioeconomic Determinants of Health and Health Economics.

Dr. Pujiyanto, MPH, is an Associate Professor from the Department of Administration and Health Policy. His Research interests are in Health Economics, Health Insurance, and Health System Policy.

Data Availability

ILAS 2023 data were publicly available at the following link: https://surveymeter.org/id/post/indonesia-longitudinal-aging-survey-ilas-2023. Data access is granted upon prior submission of a research proposal.

Declaration of Use of AI in Scientific Writing

The authors use TRINKA AI to improve lexical structure and fix grammatical errors. TRINKA does not generate ideas or produce sentences; it only corrects grammar and provides suggestions to improve paragraph cohesiveness. The authors used these tools and then conducted the necessary reviews and edits, assuming complete responsibility for the final content of the publication.